Most individuals exposed to Mycobacterium tuberculosis either never become infected or are able control infection maintaining the bacteria in a latent state walled off inside structures called granulomas. Active tuberculosis occurs in a subset of infected subjects when immunologic control breaks down. Immunologic approaches for preventing infection or disease have focussed on the critical role played by CD4+T cells and macrophages in host resistance to mycobacteria. Although the functions of these cells can be studied in peripheral blood from patients or in secondary lymphoid tissue of experimental hosts, the results obtained do not necessarily reflect immunological processes within the granuloma itself which depend both on the local tissue environment and the composition and cellular dynamics of the lesions themselves. What is needed then is to determine those immune responses that control bacterial growth within granulomas and to define the cellular interactions required for the induction and maintenance of this effector function. In collaboration with Ron Germain's lab we have developed a model system for visualizing cellular events during the formation, maintenance and dissolution of granulomas within living host tissue using high-resolution multiplex static imaging and intravital multiphoton microscopy. Because of the biosafety issues involved in working with Mycobacterium tuberculosis, we utilized avirulent Mycobacterium bovis BCG for these studies. Upon intravenous infection, this mycobacterium species forms granulomas in hepatic tissue which can be examined by multiphoton photomicroscopy in anesthetized mice in which the liver is surgically exposed. In work reported this year, we studied the initiation of the response to BCG and found that Kupffer cells in the liver directly capture blood-borne bacteria and form the nucleus of the evolving granuloma recruiting both uninfected liver-resident macrophages and blood-derived monocytes to the lesion. Within the mature granuloma, these myeloid cell populations form a relatively immobile cellular matrix that interacts with a highly dynamic effector T cell population. The efficient recruitment of these T cells was highly dependent on TNF-alpha-derived signals, which also maintained the granuloma structure through preferential effects on uninfected macrophage populations. Interestingly, we found that exogenously introduced CD4+ T cells specific for BCG or for an irrelevant Ag (ovalbumin) both migrated into the granuloma with equal efficiency demonstrating that their recruitment is not dependent on Ag recognition. Together these studies establish that modern high resolution intravital microscopy can be used to study the dynamics of myeloid, T cell interactions in mycobacterial granulomas in living mammalian tissue and provide a powerful new model system and perspective for analyzing the cellular events underlying mycobacterial containment and its breakdown in disease.[unreadable] Progress was also achieved this year in the development and characterization of an animal model for an important pathologic syndrome seen in drug treated AIDS patients. Immune Reconstitution Inflammatory Syndrome (IRIS) is the rapid and paradoxical worsening of pathology seen in HIV-1 infected individuals after initiation of anti-retroviral therapy (ART). With the increasing use of ART for mass treatment of AIDS in developing regions, this complication has emerged as a major problem in the management of HIV infection. IRIS is more common in patients with severe CD4 T cell lymphopenia and patients with M. tuberculosis or M. avium infection are at particularly high risk of developing this severe pathologic response. The factors that lead to the induction of IRIS and the mechanisms of its immunopathogenesis are poorly understood, so currently patient treatment during IRIS episodes is limited to supportive care and broad steroid based immunosuppression. We have developed a murine model that mimics the two fundamental features of the clinical scenario: profound CD4 T cell lymphopenia with concurrent mycobacterial infection. Upon initial exposure, Mycobacterium avium slowly grows and disseminates throughout the host, but both WT and TCRalpha-/- mice survive for many months following infection. However, when TCRalpha-/- mice are first infected with M. avium and then inoculated with CD4 T cells from naive mice several weeks later, they develop a rapid wasting disease in which the mice lose >30% of their body weight and succumb by day 14 of reconstitution. This pathologic response is entirely dependent on the presence of the mycobacterial infection. Moreover, TCR Tg CD4 T cells specific for irrelevant Ags do not trigger disease indicating that this is an Ag driven T cell response and not a universal property of homeostatically proliferating CD4 T cells. IFN-gamma -/- CD4 T cells induce much less weight loss than WT CD4 T cells, and neutralizing TNF delays the onset of disease consistent with its origin as a Th1 dependent response. FACS sorting of T cell subsets based on CD44 and CD25 showed that naive CD4 T cells are the most potent inducers of disease, CD44highCD25- CD4 T cells are also capable of causing disease but to a lesser extent, and that CD44highCD25high Tregs are non-pathogenic. These data suggest that in our model of mycobacteria-associated immune reconstitution disease the combination of lymphopenia and concurrent mycobacterial infection drives the reconstituting naive and pre-existing memory CD4 T cells to mount dysregulated Ag-driven Th1 responses that lead to fatal immunopathology. [unreadable] In a companion and complementing clinical study initiated with our collaborators, Irini Sereti and Mario Roederer we are using multiparameter flow cytometry to characterize T cell populations in ART treated HIV patients in order to identify phenotypic changes predictive of IRIS development.